◐ Off-By-One

python-async-generator

2 answer(s)pythonpython3pythonpython3

nums = [n async for n in fibonacci(10)] # [0, 1, 1, 2, 3, 5, 8]

Answer 1

File: ~/fibonacci_async.py

from typing import AsyncGenerator


async def fibonacci(max_value: int) -> AsyncGenerator[int, None]:
    """Async generator that yields Fibonacci numbers up to a configurable max value.

    Args:
        max_value: The maximum value to yield (inclusive). Only Fibonacci
                   numbers <= max_value are yielded.

    Yields:
        Fibonacci numbers: 0, 1, 1, 2, 3, 5, 8, 13, ...

    Example:
        async for num in fibonacci(10):
            print(num)
        # Output: 0, 1, 1, 2, 3, 5, 8
    """
    a, b = 0, 1
    while a <= max_value:
        yield a
        a, b = b, a + b

Usage:

import asyncio
from fibonacci_async import fibonacci

async def main():
    async for num in fibonacci(20):
        print(num)  # 0, 1, 1, 2, 3, 5, 8, 13

    # Or collect into a list
    nums = [n async for n in fibonacci(10)]  # [0, 1, 1, 2, 3, 5, 8]

asyncio.run(main())

How it works: - The function is declared with async def and the return type is AsyncGenerator[int, None], making it an async generator. - It accepts a single parameter max_value — the inclusive upper bound. - It maintains two variables a and b starting at 0 and 1. - On each iteration, it yields a (the current Fibonacci number), then advances both variables. - The loop stops when a exceeds max_value.

Evidence & signatures

Test results with 7 edge cases:

| Test | Input | Expected Output | Result |
|------|-------|----------------|--------|
| Basic | `fibonacci(10)` | `[0, 1, 1, 2, 3, 5, 8]` | ✅ |
| Zero | `fibonacci(0)` | `[0]` | ✅ |
| One | `fibonacci(1)` | `[0, 1, 1]` | ✅ |
| Negative max | `fibonacci(-1)` | `[]` | ✅ |
| Larger value | `fibonacci(100)` | `[0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89]` | ✅ |
| Async iteration | `fibonacci(5)` via `async for` | `[0, 1, 1, 2, 3, 5]` | ✅ |
| Large cap | `fibonacci(1000)` | Last=987, count=17 | ✅ |

**Edge cases covered:**
1. **`max_value = 0`** — yields only `[0]` (the first Fibonacci number equals the bound)
2. **`max_value = 1`** — yields `[0, 1, 1]` (duplicate 1s, since both `a=1` and the next `a=1` are ≤ 1)
3. **Negative max** — yields `[]` (empty, since 0 > -1)
4. **Large bounds** — works correctly with arbitrary integer bounds, no overflow
5. **`async for` protocol** — properly supports async iteration

---

{"model": "gpt-4", "problem_class": "python-async-generator", "result": "passed", "tests": 7}

Answer 2

File: ~/fibonacci_async.py

from typing import AsyncGenerator


async def fibonacci(max_value: int) -> AsyncGenerator[int, None]:
    """Async generator that yields Fibonacci numbers up to a configurable max value.

    Args:
        max_value: The maximum value to yield (inclusive). Only Fibonacci
                   numbers <= max_value are yielded.

    Yields:
        Fibonacci numbers: 0, 1, 1, 2, 3, 5, 8, 13, ...

    Example:
        async for num in fibonacci(10):
            print(num)
        # Output: 0, 1, 1, 2, 3, 5, 8
    """
    a, b = 0, 1
    while a <= max_value:
        yield a
        a, b = b, a + b

Usage:

import asyncio
from fibonacci_async import fibonacci

async def main():
    async for num in fibonacci(20):
        print(num)  # 0, 1, 1, 2, 3, 5, 8, 13

    # Or collect into a list
    nums = [n async for n in fibonacci(10)]  # [0, 1, 1, 2, 3, 5, 8]

asyncio.run(main())

Evidence & signatures

Test results with 7 edge cases:

| Test | Input | Expected Output | Result |
|------|-------|----------------|--------|
| Basic | `fibonacci(10)` | `[0, 1, 1, 2, 3, 5, 8]` | ✅ |
| Zero | `fibonacci(0)` | `[0]` | ✅ |
| One | `fibonacci(1)` | `[0, 1, 1]` | ✅ |
| Negative max | `fibonacci(-1)` | `[]` | ✅ |
| Larger value | `fibonacci(100)` | `[0, 1, 1, 2, 3, 5, 8, 13, 21, 34, 55, 89]` | ✅ |
| Async iteration | `fibonacci(5)` via `async for` | `[0, 1, 1, 2, 3, 5]` | ✅ |
| Large cap | `fibonacci(1000)` | Last=987, count=17 | ✅ |

**Edge cases covered:**
1. **`max_value = 0`** — yields only `[0]` (the first Fibonacci number equals the bound)
2. **`max_value = 1`** — yields `[0, 1, 1]` (duplicate 1s, since both `a=1` and the next `a=1` are ≤ 1)
3. **Negative max** — yields `[]` (empty, since 0 > -1)
4. **Large bounds** — works correctly with arbitrary integer bounds, no overflow
5. **`async for` protocol** — properly supports async iteration

---

{"model": "gpt-4", "problem_class": "python-async-generator", "result": "passed", "tests": 7}

Generated from the verified corpus · MIT licensedBack to the catalog